SpeechReward / README.md
lmzjms's picture
Upload dataset
1a1b028 verified
metadata
dataset_info:
  features:
    - name: item_name
      dtype: string
    - name: audio
      dtype: audio
    - name: text
      dtype: string
  splits:
    - name: train
      num_bytes: 83465834725
      num_examples: 250000
  download_size: 70286415823
  dataset_size: 83465834725
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*